Papers with heterogeneous graph-structured modeling of communities

1 papers
Abusive Language Detection with Graph Convolutional Networks (N19-1)

Copied to clipboard

Challenge: Existing approaches to abusive language detection only capture shallow properties of online communities . a new approach captures both the structure of online community and linguistic behavior of users .
Approach: They propose a graph convolutional network approach that captures the linguistic behavior of users . they propose to model homophily by embeddings for authors that encode the structure of their communities .
Outcome: The proposed approach captures both the structure and linguistic behavior of users in online communities . authors show that the proposed approach significantly advances the current state of the art .

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations